Modern storage tiering solutions can implement tiering at the volume level and usually within large storage servers using large chunks of data (256 MB and up). Due to the size of these chunks of data, a chunk can contain a large variety of data that can affect the overall usage of a block. The chunk can contain some data sets that are not highly accessed. For data sets that are not highly accessed, the access is affected because the entire block remains at a lower tier, when some of the data should be at a higher tier.